22

There are two closely related questions here:

  1. If a student submits their assignment and is unhappy with their mark, are there any dangers (which I might not be seeing) in allowing the students to do so? If I would offer for one, I would offer for all, but should I be offering it in the first place?

  2. If resubmits are unavailable, some students ask me to 'pre-read' or 'pre-mark' their assignment. The end result is that when they submit officially, that may actually be the second submission. Clearly doing any 'pre-marking' or the like takes time and energy and, so far, very few students have taken advantage of this until right before the submission deadline (at which point I stop because I don't have THAT much time to offer them). Are there dangers in 'pre-marking' assignments?

Note: These assignments constitute either 50% or 100% of their total mark.

One of my concerns is that the student would have lower motivation to maximize the quality of their original work and just correct what was marked as a problem area - like a production worker depending on a Quality Control Inspector and not paying as much attention to the quality of their work the first time around. Are there other issues as well?

EDIT: I should add that these are business management subjects so student answers are not easily right or wrong but more about how they justify their analysis. Therefore, there is not an issue of "giving the right answer."

  • 1
    You should also ask about the danger involved in not allowing this, since it's pretty significant. – Brendan Long Sep 9 '13 at 19:21
  • @BrendanLong I asked based on the assumption that pre-marking and multiple submissions of the same work is not common amongst teachers. However, I would be interested to hear the other side (of not allowing this) – earthling Sep 10 '13 at 0:01
  • The danger of not allowing retries is that you perpetuate a system where students are taught that if they fail at something, they should give up. In the real world, the most successful people are the ones who take risks, and understand that failure is just part of the learning process. At my job, failure isn't considered a problem unless I failed and didn't learn something from it. – Brendan Long Sep 10 '13 at 0:17
  • @BrendanLong I'm not sure about this. The idea has always been you do your best. If you don't achieve all you hoped to then take the lessons learned and apply them to do better next time. We don't get several chances to be first to market. We do get several chances to succeed overall in business and students also get several assignments to pass a module. One poor assignment does not condemn them to failing that module. – earthling Sep 10 '13 at 4:54
  • Why would you want to do this? – StrongBad Sep 10 '13 at 11:41
14

Should I be offering it in the first place?

Dangers:

  1. You spend even more time on assignments, limiting your time for other endeavors.
  2. Students might get into a "grade grubbing" mode where they simply re-submit marginally better answers in order to improve their grade.
  3. Students do a poor job initially because they know they will have a chance to re-submit, and they might as well take their chances that you'll give them decent marks for inferior work.

Benefits:

  1. Overall, students spend more time on the assignments, leading to better knowledge and ability to do the work.
  2. You can provide helpful information to guide them to better answers (because you're seeing first-attempts). This level of grading also takes more time.
  3. Students are happier because the stress of a hard deadline isn't so bad.

I think the benefits outweigh the dangers if you're willing to put in the extra time to re-grade. As I've said in other threads, I've used auto-grading homework for a some classes (probably not relevant in your case), but I've used the strategy that I'd have deadlines for all the assignments but that the week before the final I'd re-open all the assignments and tell the students they can re-do any questions they missed. I don't tell them I'm going to do this until I do it, in order to keep students from simply waiting until the end to do all the assignments.

Are there dangers in 'pre-marking' assignments

I think it's great that you're being creative with the assignments (e.g., pre-marking, resubmittals, etc.). Again, it comes down to your time -- if you have motivated students who want pre-marking and you have the time, I can't see a problem with it. It wouldn't surprise me if the best students are the ones who want pre-marked work, but then again I've had some less-capable students jump at chances like that to do better.

  • 19
    I had a professor that would let us resubmit our tests (one time), but for only half of the credit back. So if you got a 50 on the test initially, you could only get a 75 if you corrected it to a perfect submission. I feel that this could alleviate some of the risk with your third danger because students who do poorly on the first try are more limited by what they can get back. Additionally, the one re-submit limit helps with spending too much time grading and any grade grubbing. The problem with this class was that it was always necessary to resubmit because class average was around 30-40%. – Gray Sep 9 '13 at 12:27
  • 1
    @Gray I actually prefer that method; I wrote an answer around the same time as you commented. – Amory Sep 9 '13 at 12:31
9

A suggestion that I have used quite successfully in regards to your point no. 2: premarking. This would depend on the dynamics of your class, but if at all possible to have the students perform a peer-review pre assessment of each other's work.

With peer-reviews, the students become more aware of is needed to fulfil the criteria and requirements. Alongside that, each student can also do a self review with a criteria based self-review pre-assessment.

There is always a risk of copying, but I have found that this to be no more than what would occur with the normal procedure (and tools such as turnitin can help with this).

  • 3
    +1 for peer review. If you can get it to work in your class (and you have time in the curriculum for it), that is a terrific option. – Chris Gregg Sep 9 '13 at 8:49
  • +1: peer review is an excellent exercise in its own right. Reading with a critical eye is as important a skill as writing well. – evilcandybag Sep 9 '13 at 13:20
  • 3
    I've found that peer review isn't very helpful. Other students usually aren't good enough to provide useful feedback. I saw this all the time in writing classes, and the peer review was always useless. – Brendan Long Sep 9 '13 at 19:23
8

It's certainly tricky! There are a number of pitfalls, I've found, that usually end up making it not worth the effort:

  • A lot more work for you. Grading twice may be no biggie in a class of 10, but a class size of 200 or 300 would be prohibitive.
  • The students will, in all likelihood, not try as hard the first time around, since they know they automatically get an extension.
  • Grade inflation. This may not be a big deal to some (it's a minor crusade of mine) but it can be a major issue. If the entire class is getting >90%, your superiors may be suspicious.
  • Similarly, the more successful kids are liable to get annoyed. What's the point of busting your hump to be an A-student if the B- and C-students can resubmit and do just as well?

That being said, there are some good work-arounds; nothing you can do about class size, but here are some ideas:

  • Don't tell them beforehand! This is a bit devious, and you can only do it once and not on the final, but most people will, after a hard midterm or something, appreciate that hugely. This will ensure most everyone works their butt off instead of slacking first time around.
  • Give them half credit for resubmitted answers. That is, if a student gets an 80%, but the resubmit is 100%, give him/her a 90%. This can be a good (enough) balance: The poorly-performing students get a sufficient boost, but not enough to threaten or annoy the high-performers. This also reduces the overall grade-inflation risk.
  • In the vein of pre-marking, since there's no "right answer," you can accept "drafts" and then spend some time going over some of the more and less successful avenues students chose. Everyone may end up doing something similar, but students will end up thinking more about their answers.
  • 4
    I think the half-credit method can work well. Like I mentioned in my other comment, I think this introduces some new dangers. The tests/assignments shouldn't be written in such a way that a student cannot reasonably expect to pass it on the first try. This could result in (at least the appearance) of lazy teaching - couldn't prepare effectively for attempt 1, so they make it up after they know exactly what to study. The other problem is that the workload for your particular class becomes higher since students have to do each test/assignment twice - something I think a lot of teachers forget. – Gray Sep 9 '13 at 12:38
  • Couldn't agree more. – Amory Sep 9 '13 at 12:39
  • I had a class where this was generalized, so that when re-grading, you lost 10% per re-grading (or day?), so you could submit a project 8 times, but by the 8th time, you can only get 20% of the points back. It worked well in that class because it was programming, so re-grading takes almost no effort, but I think it's worthwhile to allow as many re-grading attempts as you're willing to put time into. The result of this policy was that people did really well in that class, because people put in the effort to make their programs good instead of good-enough. – Brendan Long Sep 9 '13 at 19:30
7

Since the field you're in is different from mine (mathematics), what I am about to describe may not work for you, but perhaps some variant of it could work.

I am quite fond of allowing students to make corrections to assignments (especially quizzes and exams), with the following rules:

1) Students may only resubmit an assignment once.

2) The corrected solutions must be flawless. There is no partial credit on resubmissions.

3) The students can earn back half of the points that they lost on their first submission. So if they scored a 4/10 on a problem the first time around, and if they submit a flawless corrected version, they will improve this to a 7/10. They cannot do any better, and there is nothing in between.

It has been my experience that this is a great way to get students to correct and learn from their mistakes, and I cannot find a downside to it. After all, my goal is to get the students to learn the material!

Since your assignments do not have clear-cut answers (as you describe in your edit), this might need some tweaking. You may need to replace "flawless" with "would have earned very high marks" or something to that effect. But the point remains: I think that allowing students to resubmit assignments is a valuable tool.

6

Occasionally, students will ask me to pregrade or review an assignment prior to submission. I am always happy to do so, but I don't give them a grade, just feedback on what is good, what is lacking, and what needs improvement. I find this is a good compromise.

4

I think the problem lies in what expectations the students have. With a clear set of grading criteria (c.f. rubrics) you could probably avoid some or many problems. In the second case, you could say that pre-grading provides a preliminary grade which. at best, will be raised by one notch upon successful resubmission. Then, the student will get feedback and know where they are and make a decision of putting the effort it worth it. Clearly, this simple example is not thought through so I believe this case may need a serious think and discussion with peers on possible side effects. But, the point is that if you make the grading clear then there will be less complaints.

Your first case is perhaps trickier. The same idea applies, you need to apply some clear rule of what happens with a resubmission. It is possible that if your design of the criteria for case 2 is good, then as a side-effect case 1 disappears. You should also consider reducing grades for late arrivals etc.

Although this reply is not very precise, I think setting up criteria that makes the grading process clear will help. By putting in an optional pre-assessment, you make the students decide what they should do. Note that the criteria for the assignment also have to be clear so that the grades you set can be explained in terms of learning objectives.

  • 1
    I agree with Peter about setting up clear criteria -- whether that is, for instance, only letting students gain a certain percentage more for the assignment, or otherwise, having something you can point to if students try to go beyond the parameters will help. – Chris Gregg Sep 9 '13 at 8:30
  • Thanks for these thoughts. I actually provide a rubric but some (the weaker) students have difficulty marking their own paper. – earthling Sep 10 '13 at 0:08
4

One advantage of allowing resubmits that hasn't been pointed out is that it allows you assign more challenging material without scaring the students too much.

3

Regarding the first scenario: if you give out solutions, there's obviously a problem. But if you hold off on giving out solutions, then the remaining students have to wait longer than might be reasonable.

  • 1
    This made me realize I skipped something important. I've edited the question accordingly. – earthling Sep 9 '13 at 6:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.